错误：AttributeError: 'ChatGLMTokenizer' object has no attribute 'tokenizer'

[longmarchw]

执行：

import torch from modelscope import AutoModel, AutoTokenizer

model_id = 'ZhipuAI/codegeex2-6b' tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True) model = AutoModel.from_pretrained(model_id, device_map={'': 'cuda:0'}, # auto torch_dtype=torch.bfloat16, trust_remote_code=True) model = model.eval()

prompt = "# language: python\n# write a bubble sort function\n" inputs = tokenizer.encode(prompt, return_tensors="pt").to(model.device) outputs = model.generate(inputs, max_length=256) response = tokenizer.decode(outputs[0]) print(response)

----------报错-------- AttributeError: 'ChatGLMTokenizer' object has no attribute 'tokenizer'

[zhou20120904]

同错

[longmarchw]

我已经解决了。是版本对不上，需要安装 `pip install cpm_kernels protobuf==4.24.4 "transformers<4.34"，默认transformers版本高了。

[Songzi1228]

transfomers==4.32.0，还是这个错误

[thewangcj]

transfomers==4.32.0，还是这个错误

pip install transformers==4.30.2，requirements.txt 中有

[NytePlus]

transformers==4.30.2, protobuf==4.24.4, cpm-kernels==1.0.11还是这个错误

[NytePlus]

transformers==4.30.2, protobuf==4.24.4, cpm-kernels==1.0.11还是这个错误

抱歉我错误地配置环境变量导致jupyter notebook的kernel显示为配置好的conda环境，但实际上sys.path是其他。这个版本对我来说可以解决问题